VTE events in TBI remain a major problem with a significant impact on complication rates, and mortality. A recently published systematic review showed that 20% of patients with isolated TBI had laboratory coagulopathy on hospital admission [16]. However, the effect of TBI on coagulation is not well understood, and in particular, the role of the mechanism of injury with respect to VTE events has not been clinically determined.

Our hypothesis that penetrating compared to blunt trauma mechanism is independently associated with more VTE events in TBI was confirmed. This is an interesting finding that, to our knowledge, has not been described previously.

To understand the role of the mechanism of injury in VTE complications, it is important to be familiar with the underlying interactions of the coagulation cascade that may contribute to VTE complications in trauma and particularly in TBI. In an early phase after trauma underlying mechanisms such as platelet dysfunction [17], hypoperfusion induced activation of protein C may contribute to a hypocoagulable state [18]. In the further course, hypocoagulability seems to transform to a hypercoagulable state with an increased risk of thrombus formation. In this phase the release of tissue factors from local tissue injury, as well as increased systemic production may activate the extrinsic clotting cascade [19, 20]. These findings support the clinical hypothesis that tissue factors drive hypercoagulability after brain injury. Another study by Meizoso et al. [21] recently measured thrombelastography indices in trauma patients with and without TBI. The study provided evidence that fibrinolysis shutdown associated with a hypercoagulable state was more common in patients with TBI as compared to patients with no TBI (25% vs 18%, p < 0.0001). Finally, it is discussed that in shock, massive consumption of protein C and Interleukin 6 secretion may also play a role in thrombus formation [22, 23]. Only one study by Martin et al. [11] compared blunt vs penetrating TBI patients with regard to coagulopathy [measured by thromboelastography (TEG) and need for transfusion]. Patients with penetrating head injuries were more coagulopathic by TEG and were also more likely to undergo transfusion with any type of blood product compared to those with blunt injuries (26.5% vs 6%, p < 0.0001). Unfortunately, no VTE complications were reported in this study. Nevertheless, it is likely that the group of patients with penetrating injuries who initially presented with accentuated coagulopathy compared to blunt trauma patients, may also have experienced an increased prothrombotic state during the clinical course.

Most clinical research on VTE complications in TBI has focused on blunt trauma. Only a limited number of studies included patients with penetrating injuries. In a retrospective database review of a general trauma population, rates of VTE were the same for blunt and penetrating mechanism [24]. However, in another study from 2021 [25] firearm injury was identified as the strongest mechanism of injury associated with VTE complications in a general trauma population (OR 1.88, CI 95% 1.39–2.54; reference: other injury).

Another study by Meyer et al. [26] was evaluating VTE chemoprophylaxis in penetrating TBI. Despite very early VTE chemoprophylaxis (within 24 h) the study reported VTE rates of 12% in combat-related penetrating brain injury. This high rate may support our findings identifying penetrating injury mechanism as an independent risk factor for VTE complications in TBI.

The decision to start VTE prophylaxis in TBI is extremely challenging: the initiation of VTE prophylaxis too late may result in thrombosis; whereas the initiation of premature prophylaxis may increase the risk of progression of intracranial bleeding, especially in severe TBI [27]. The American College of Surgeons state in its best practice guidelines for the management of TBI that in most cases, VTE prophylaxis should be considered within the first 72 h after TBI [28]. However, several recently published studies evaluated a more aggressive VTE prophylaxis in TBI patients and concluded that VTE prophylaxis within 48 h of admission is safe and effective in preventing VTE complications in TBI patients [29,30,31,32,33]. As a consequence of the increased VTE risk in penetrating as compared to blunt TBI an even more aggressive VTE prophylaxis might be appropriate. In the study by Meyer et al. [26] evaluating penetrating TBI the very early administration of VTE prophylaxis (within 24 h) was safe with regard to the progression of intracranial hemorrhage. However, more high quality data are needed to provide sufficient evidence to support this aggressive VTE prophylaxis management in penetrating TBI.

The underlying mechanisms for higher rates of VTE events in penetrating compared to blunt TBI in our study could not be examined. Higher release of procoagulants (tissue factors) or an increased fibrinolysis shutdown in penetrating compared to blunt trauma would be a possible explanation but is only hypothesized [10]. Further studies should focus on a better understanding of the effects of the mechanism of injury on the coagulation cascade in TBI. This may help to further improve risk stratification regarding the initiation of VTE prophylaxis.

In addition to the penetrating mechanism of injury, the present study identified the following risk factors for VTE events in isolated severe TBI: increasing age, male gender, obesity, tachycardia, increasing head AIS, associated moderate abdominal, spinal, upper or lower extremity injuries defined as AIS 2, neurosurgical procedures (craniectomy/craniotomy or ICP monitoring) and pre-existing hypertension. Increasing GCS, early VTE prophylaxis and LMWH compared to heparin were identified as protective factors.

A study from 2019 using a national registry also reported on risk factors for VTE following TBI [34]. This study retrospectively evaluated 424,929 patients with TBI between 2002 and 2014. The overall described VTE rate was 3.9% compared to 2.6% reported in our study. The higher VTE rate compared to our study may be explained by the fact that patients who died within 72 h were not excluded in this study. These are likely the most severely injured patients with the highest risk for VTE complications. In addition, the study included patients with severe associated injuries which also may have contributed to higher VTE rates. However, in line with our findings the study identified older age, hypertension, obesity, neurosurgical procedures (craniotomy/craniectomy, EVD or ICP monitor) and more severe TBI as independent risk factors for VTE complications. In comparison to our study which showed an independent trend toward an increased mortality for pulmonary embolism [1.35 (95% CI = 0.97–1.87)], the study by Hoffman et al. identified VTE as a protective factor for mortality [0.53 (95% CI = 0.50–0.57)]. This counterintuitive finding may be explained by confounding factors that were not considered in the study design; or may also be explained by the fact that patients who died had a shorter length of stay and therefore had less time to develop a DVT or PE. The following limitations of the study by Hoffman et al. must be taken into account: First, associated injuries were neither reported nor included in the regression analysis. Our study showed that even moderate extracranial injuries (AIS = 2) of the abdomen, spine, and lower/ upper extremities were independently associated with VTE complications in patients with TBI. Second, duration of consciousness was used as a surrogate measure of the severity of TBI because GCS was not coded in the database. However, unconsciousness, for example, is also induced by severe hemorrhagic shock and is therefore an unreliable method for assessing the severity of TBI. Lastly, the study by Hoffman et al. did not consider trauma mechanism (blunt vs penetrating), VTE type and timing in the regression analysis. Not taking these factors into account including the inaccurate surrogate marker for severity of TBI may have confounded the results.

Another large retrospective database study using the National Inpatient Sample database evaluated in 2022 risk factors for VTE complications in adult patients with severe TBI [35]. This study evaluated 349,165 TBI hospitalizations. In line with our findings, the study found an independent association between craniectomy and an increased risk of VTE for patients with severe TBI (OR 1.29, p < 0.005). Furthermore age (OR 1.26, p < 0.005), chronic lung disease (OR 1.58, p < 0.05), electrolyte imbalance (OR 1.43, p < 0.05), liver disease (OR 0.10, p < 0.05), urinary tract infection (OR 1.56, p < 0.05), pneumonia (OR, 2.03, p < 0.0001), and sepsis (OR 1.57, p < 0.05) were also identified as independent factors associated with VTE events. Obesity (OR, 2.09, p > 0.05) and spine injury (OR 2.03; p > 0.05) showed a trend toward an increased independent risk for VTE events. In this study, associated severe injuries were also not excluded, and the regression analysis corrected only for spinal injuries and leg or pelvic fractures. On the other hand, complications such as sepsis, pneumonia, urinary tract infections, acute myocardial infarction, or cardiac arrest were included in the regression analysis. In our opinion, the inclusion of complications as predictive variables for VTE events is of limited use, because the mentioned complications develop during the clinical course and hardly contribute to the necessary early risk stratification for thromboembolic events.

To our knowledge, this is the first study identifying penetrating trauma mechanism as an independent risk factor for VTE events in isolated severe TBI. Definitely a strength of this study includes the large number of patients, the quality of the TQIP databank, and the evaluation of patients with isolated severe TBI. This helps minimize confounding effects of severe non-TBI injuries with regard to VTE events. In addition, to assess a homogenous population, only patients receiving VTE prophylaxis with either unfractionated heparin or low-molecular-weight heparin were included in the analysis. Furthermore, a broad number of potential confounders were considered for analyses and the risk factors for VTE events in isolated severe head injury were tested in an additional case control matching analysis.

However, there are a number of limitations and our results should be interpreted with caution. First this is a retrospective study, based on a large database and it is therefore associated with all the inherent limitations of this study design. The strength to evaluate only patients with isolated severe TBI is also a limitation because patients with TBI often present with severe concomitant injuries. As our results showed, even moderate associated injuries (AIS = 2) increased the risk of VTE complications. Therefore, it is very likely that VTE rates are even higher in polytraumatized patients with combined head and associated injuries. In addition, doses, duration and held doses after initiation of VTE prophylaxis, including clotting tests, are not recorded in the TQIP database and could not be considered for analysis. Finally, the present study did not examine specific mechanisms of injury (gunshot or stab wound for penetrating injury). This could be the focus of future studies.